Contents - Previous - Next


2.2 History and organization of moving image archives

Sam Kula

The Archival appraisal of moving images: a RAMP study with guidelines. 1983

Editors note: Although this paper was written some 14 years ago many of the points are still relevant and the paper is included as a history of the development of moving image archives.

1.1. That moving images of actual people, places and events - actualities, as they were termed at the turn of the century - are historical documents with unique properties was being recognized as early as 1898. Just two years after the first public exhibitions of cinematography in Paris, London, Berlin and New York, Boleslav Matuzewski, a Polish cinematographer in the employ of Nicholas II of Russia, published a manifesto in Paris calling for the establishment of a world-wide network of archives to acquire and conserve the product of this new marvel of technology, this new source of history. Matuzewski was aware that for cinematography to fulfil its historic mission it would first have to move from purely recreational or fantastic subjects toward actions and events of documentary interest; from the slice of life as human interest to the slice of life as the cross-section of a nation and a people". In other words the objectives of the cinematographers would have to differ from those of their predecessors who had developed the magic lantern and optical toys, based on the persistence of vision and the illusion of motion, to a very sophisticated level as a means of personal or public amusement.

1.2. It is interesting to note that Matuzewski also foresaw the eventual need for the appraisal and selection of moving images. One of the functions of the "competent committee" he proposed to administer the 'Depository of Historical Cinematography' to be established in Paris was to "accept or reject the proposed documents according to their historic value". He assumed that the volume of voluntary deposits by cinematographers anxious to have their films permanently conserved would demand an appraisal policy.

1.3. Matuzewski was not alone in discerning the recording properties of the motion picture camera. In 1900 the Ethnographic Congress in Paris proposed and adopted the resolution: All anthropological museums should add suitable film archives to their collections. The mere possession of a potter's wheel, a number of weapons or a primitive loom is not sufficient for a full understanding of their functional use; this can only be handed down to posterity by means of precise cinematographic records.

Anthropologists and ethnographers were thus early to begin utilizing the new technology as an additional tool along with the still camera and the audio recorder, but there was little systematic coverage of traditional cultures undergoing rapid change. As Francis Speed points out in describing the situation in Nigeria, the irony is that this product of the new technology has not been more widely employed in recording cultures threatened by the introduction of all types of new technology. "Already many of these cultures have disappeared; their ways of life have basically and irrevocably changed and comparatively few have been reliably and comprehensively documented.

. . .In a living culture the majority of the ritual, social, political and working occupations entail continuous and complex movements. In order to make comprehensive and reliable records of these activities it is necessary to use an audiovisual medium capable of reproducing movement".

1.4. Despite the enthusiasms of the few who perceived the new medium of motion pictures to be capable of capturing a unique record of human endeavour in the twentieth century, and despite the almost universal public acceptance of motion pictures - the first mass medium to transcend both international boundaries and cultural differences - the moving images produced were almost totally neglected by librarians, museum curators and archivists. In the beginning there were essentially two approaches to film-making as far as the general public was concerned: the first were those who, following in the footsteps of Auguste and Louis Lumière, found their subject matter in the real world around them; the second, inspired by the fantasies of George Méliès, an illusionist turned film-maker, discovered that the limitations of the real world in terms of time and space could be overcome through the "magic of the movies", the transformation of real time and real space into screen time and screen space. The initial objectives of both approaches in the commercial cinema was to entertain, to attract an audience, to the profit of the producer, the distributor and the exhibitor. The followers of Mé1iès proved to be more successful at the box-office, especially after the development of narrative techniques, and the actualities began to play a minor role in the emerging film industry.

1.5. Moving images quickly became associated with vulgar entertainment and their usefulness was measured by the financial return they could generate. Although in numbers well over half the moving images produced during the first quarter of the twentieth century continued to be actualities, the feature length fiction film exemplified by the standard productions of the major Hollywood studios ("the dream factory") so dominated the public consciousness that all moving images were regarded by the custodians of artifact and culture as escapist fare of no lasting value. Even as moving images in the twenties and thirties began to be harnessed in the service of national and international ideologies, and as the impact of moving images as shapers of public opinion and moulders of public taste began to be recognized by politicians and advertisers alike, there was no concerted effort to systematically acquire and conserve the moving images of one generation for the enjoyment and edification of those to follow. As a result it is estimated that fully one-half of the moving images produced before 1930 have been lost.

1.6. One of the major contributing factors was the nature of the nitrocellulose stock which was used for all theatrical productions prior to 1950. Although long wearing and with excellent optical qualities, nitrocellulose is an inherently unstable compound that gradually but inevitably disintegrates over time, even under optimal storage conditions. All productions on 35mm film, the industrial standard, made before 1950 were on nitrocellulose or 'nitrate' stock.

1.7. The only method of conserving moving images on nitrocellulose stock prior to 1950 was to transfer the image to fresh stock, and the prospect of having to do this every few years, along with the storage problems associated with highly flammable materials, acted as a powerful deterrent on archives, libraries and museums that might otherwise have assumed responsibility for what was becoming recognized as a vital part of the public record. The Library of Congress in the U.S., for example, accepted the deposit of moving images as series of photographs printed on paper rolls for the purpose of copyright registration from 1896 through 1912 and thus acquired an invaluable collection of more than 3,000 pioneer moving image productions. However, when new legislation on copyright permitted the deposit of nitrocellulose stock prints, the Library changed its procedures so that presentation of a copy could serve as evidence for copyright registration in place of the actual deposit. The Library did not resume the acquisition of moving images (with rare exception) until the forties, and did not develop a full scale acquisition and conservation program until the end of the sixties.

1.8. It was not until the early thirties, when the introduction of sound had placed all silent films in danger because they were no longer commercially competitive, that the first archives specifically devoted to the acquisition and conservation of moving images were organized. These first archives were founded on the work of individuals, and frequently were based on private collection that were "institutionalized" in order to ensure funds for conservation, control over access, and the continuity of support necessary to enlist the cooperation of depositors. Perhaps the most influential pioneers were Henri Langlois (Cinémathèque Française, Paris), Ernest Lindgren (National Film Archive, London), and Iris Barry (Museum of Modern Art, New York). Through the force of their personalities they secured public recognition of the need for moving image archives, and established the legitimacy of their calling.

1.9. In 1938 four of the pioneer organizations, the Cinémathèque Française, the Museum of Modern Art Film Library, The National Film Archive, and the Reichsfilmarchiv, Berlin, founded the International Federation of Film Archives (FIAF). World War II interrupted that development, but the Federation was re-established in 1946 by the archives in Paris, New York and London, with the addition of Gosfilmofond, Moscow. The Federation had 46 members and 23 observers from 52 countries in 1982. Although the majority of the members are from Europe and North America, all continents are represented, and it is one of the Federations' chief objectives to assist in the formation and growth of moving image archives in developing countries.

1.10. An analysis of FIAF's (International Federation of Film Archives) membership reveals that very few moving image collections have been established in national archives or libraries. The collections at the Public Archives of Canada, the Library of Congress (U.S.), the National Library of Australia and the Bundesarchiv (FRG), are notable exceptions. The Federation's Statutes that demand an autonomous structure for the moving image activity as a condition for membership may have deterred other national archives or libraries that do acquire moving images from seeking affiliation, but the National Archives (U.S.) is perhaps the only such organization with a significant collection that is not a member.

1.11. The vast majority of FIAF members and observers in Western Europe, North America and Latin American are autonomous, private organizations, or affiliated with film institutes or film schools. They usually receive governmental financial support directly or indirectly but their policies and programs are developed and implemented by a small professional staff responsible to some type of governing board of directors.

1.12. In countries where the production of moving images is a state monopoly most of the FIAF members and observers are organizationally linked with the motion picture industry, usually in a structure where they serve as the archives of the production components, and as a resource base for the film institutes or schools that carry out the training programs for film makers.

1.13. With rare exceptions (Motion Picture, Broadcasting and Recorded Sound Division of the Library of Congress, Washington; National Film, Television and Sound Archives Division of the Public Archives of Canada, Ottawa; and National Film Archive of the British Film Institute, London) film archives do not systematically acquire moving images from television, either as film or videotape. Because so much of television broadcasting as introduced in the early fifties, was "live", there was no permanent record to select and acquire, and the film recordings (kinescopes) that were manufactured to allow the re-transmission of certain broadcasts at a more convenient time were either ignored or allowed to accumulate in the offices of the broadcasters.

1.14. As broadcast television became a great consumer of film, either produced for television or purchased for broadcast, more and more newsfilm, documentaries, and fiction films ended up on the shelves of the networks or in local station libraries, usually closed to the public and frequently neglected in terms of organization and conservation so that the material could not even serve effectively as a resource for future production.

1.15. With the introduction of videotape in the early sixties a record of what was broadcast could be retained for archive purposes, but in fact very few broadcasts left the hands of the producers for research or record purposes. Worse still the cost of the raw stock videotape and the fact that it could be erased and re-used combined to effect the loss of thousands of hours of programming. In short, with regard to archives, the first twenty-five years of television broadcasting throughout the world replicated the dismal history of film-making in the first forty years. The documentation, when it survived, remained in the hands of the producers and distributors, whose mandate seldom included conservation or organization for public access.

1.16. By the mid-seventies, however, the value of television broadcasts as a future production resource was becoming well established, and the importance of television broadcasting as an integral part of the public record was increasingly recognized by researchers in many disciplines. The response from the television producers was to re-evaluate and in many cases to reorganize their production resources as archives, accepting the fact that even internally the programmes and programme elements had to be protected from indiscriminate and irresponsible use due to the always pressing demands of the broadcast schedule. At the same time academically orientated bodies such as the International Association for Audio-Visual Media in Historical Research and Education, and national organizations such as the Television Archives Advisory Committee (US), the Association for the Study of Canadian Radio and Television (Canada), and the British Universities Film Council (UK), began focussing attention on the need to conserve the record of television broadcasting and to organize the resources for at least limited public access by researchers.

1.17. In 1978, primarily through the initiative of Institut National de L'Audiovisuel (Paris), the British Broadcasting Corporation (London), Radiotelevisione Italiana (Rome), and Norddeutsher Rundfunk - Fernsehen (Hamburg), the archives of the major television networks throughout the world established the International Federation of Television Archives (FIAT). Membership at present is restricted to archives of production organizations, or to those media archives that have been officially designated as the archives of a television network or production company, such as the National Film Archive (London) for Independent Television Authority companies, and National Film, Television and Sound Archives (Ottawa) for Global Television. As more national archival organizations have acquired responsibility for the conservation of moving images from television the Federation is considering modification of its Statutes to allow their affiliation.

1.18. There were only 33 members in FIAT at the end of 1982 with the membership concentrated in Western Europe. The conditions that prevail in almost all other state television networks (the norm for television broadcasting through the world) is similar, in that archives, if they exist as functional entities, remain the responsibility of the producer. In rare cases this responsibility has been delegated to a national library (as in Sweden and the U.S.), but this development is so recent that it is too early to say whether these programs will become models for other countries.

1.19. With the exceptions noted above, the involvement of national archives in the selection and conservation of television broadcasts is minimal. Substantially less, in fact, than with moving image documents from other sources, although the technology associated with conservation and public service on videotape materials is actually less complicated and less costly than it is with early film.

1.20. The deterrent in this case is volume. It is not unusual for a single broadcaster to generate over five thousand hours of programming per year. Selection is obviously essential, and what is needed, in effect, is a records management approach that will ensure the immediate protection of all the records generated for a limited time, to allow time for an evaluation of the total production for archival purposes and the preparation of a schedule which will specify which programmes are to be retained for long term conservation. This is the stated objective of the network archives that are members of FIAT.

1.21. The Institut National de L'Audiovisuel has established such as program for television in France, but the norm for television archives that are not actually part of the production organization is a much more selective approach based on advisory committees and consultation with subject experts. Since so much of television's output consists of films which may also have been or will be available through theatrical distribution, there is obviously a large potential overlap with the work of film archives. Television world-wide is also a rapacious consumer of theatrical feature films and documentaries. The separation of moving images by delivery mechanisms (television versus theatrical distribution) for archives purposes may thus be wasteful of public resources, with the potential for dividing the work of one image maker between two archives that may well be in different cities.

1.22. The degree to which archives in television networks can serve and should serve the general public of scholars and researchers is also an issue. Indicative of the nature of the problem is the Report of the Advisory Committee on Archives, established by the British Broadcasting Corporation (BBC) in 1978.

Although recommending that the BBC's archives should be made more accessible to outside researchers and the general public through in-house services and various diffusion policies, the Report recognizes that additional funding will have to be found or allocated to allow the archives to both serve the network and this wider public. One of the key elements in any such diffusion program is the computerization of data now in card catalogues so that information on the holdings of the network could be available in educational establishments throughout the country.

1.23. In recent years, as the tape recorder became a ubiquitous witness to public events and private conversations, archives at the national, regional and local level have been increasingly acquiring recorded sound as a component of other accessions, whether public records or private papers. In much the same way film and video recordings are proliferating in archival record or manuscript groups. It is probably safe to say that the annual accession list of every national archives would show some intake of moving image and recorded sound, whether or not these organizations subscribe to the theory of "the global archives", and systematically acquire and conserve this type of documentation.

1.24. How this type of documentation is treated, however, once it enters the archival system, varies widely. The range is from a policy that can be termed 'benign neglect', the passive registration and shelving of the object along with the textual material in the same group or collection, to an active policy of conservation and public service that not only recognizes that such documentation must be segregated physically to protect the recordings and, in the case of film on nitrocellulose stock, the repository itself, but also that such documentation must be inventoried, or described in greater detail than is probably the norm in a national archives to make them readily accessible to researchers.

1.25. Despite the best efforts of Commissions in FIAF and FIAT in attempting to standardize the cataloguing of moving images, little has been accomplished beyond some movement toward a definition of terminology and an acceptable list of minimal data elements. These organizations, together with the International Federation of Library Associations (IFLA) and the International Association of Sound Archives (IASA) are currently collaborating in a joint effort at standardization. In practise archivists in production organizations, such as television networks, tend to more detail in describing their holdings in order to meet the needs of their colleagues in production who may request highly specific shots or sequences. This approach is beyond the capacity of more general collections - in fact television archives tend to reserve this treatment for newsfilm, - but there is little general agreement as to what constitutes an acceptable compromise. When one considers that one minute of a motion picture, measured at sound speed, contains 1440 photographs, some of the problems involved can be appreciated. In order to cope with the volume of data generated through this type of analysis, many of the television archives have already adopted machine readable cataloguing systems, or are experimenting with computerized retrieval systems. The facility to rapidly exchange data on holdings should accelerate the move toward standardization.

1.26. Among the fundamentals in the archival handling of moving image documentation is the fact that all reference to the material demands a mechanical and/or electro-optical reproduction that inevitably wears out the copy being viewed or auditioned. It is thus at once both difficult to scan a large quantity of audiovisual documentation and costly in terms of the manufacture of reference material and the reproductive machinery required. As the viewings must frequently be supervised, although the advent of the videocassette has minimized this need, the cost in human resources is also significant.

1.27. If the master recordings are not protected by reference copies, the material on deposit is not being conserved. Even restricted and supervised access to such material is dangerous as the number of times a film, videotape or audio tape can be played before the quality of the recording is affected is very limited, and there is always the possibility of accidental damage in any pass through a piece of machinery. Staff responsible for organizing, describing and controlling access to such material must therefore have special training that includes some familiarity with the technology involved.

1.28. The material itself should be stored under environmental controls that reduce the hazards of airborne contaminents, fluctuations in humidity, and meet the needs of the base material - nitrocellulose and triacetate stock, videotape fabricated from a range of polymers with regard to temperature. In addition colour film emulsions composed of fugitive dyes, the industry standard until 1980 (colour print stocks have recently been introduced with superior keeping characteristics), will require cold storage at or below the freezing mark if the colour is to be retained more than six or seven years.

1.29. Staff expertise is also required to establish and to maintain a systematic acquisition program from major producers, whether private corporations or state monopolies. Not only will those sources demand access to the material deposited, but they will require access to copies for further production purposes that meet the standards of the industry. The archivists thus have to balance their primary objective of protecting the original material with the need to satisfy the donor a requirement which is probably contractural and which is essential in any case if further deposits are to be forthcoming. In a Unesco survey conducted in 1973 only twelve archives enjoyed any form of legal deposit with regard to moving images, and ten of those were in countries where production of moving images is a state monopoly. Voluntary compliance is still the norm in building archival collections of moving images and recorded sound, and the archives must be responsive to donor needs, particularly if a program of record management is to be introduced which guarantees the protection of material identified as archival while it still rests in the hands of the producers and distributors. (This situation is changing in most cases for the better as Section I: 1.8 and 1.9 will provide more detail).

1.30. This very sketchy survey of moving images and recorded sound in archives has attempted to indicate that almost all of the activity to date has taken place in non-governmental archives; that historically there have been physical and technological, and to some extent prejudicial factors at work to limit the involvement of governmental archives; and that the archival involvement demands physical segregation of the material in environmentally controlled stores and staff expertise in the processing and custody of such materials. The Unesco Recommendation concerning the safeguarding and preservation of moving images as adopted by the General Assembly in Belgrade, October, 1980, called for the establishment of officially recognized archives "by each member state to acquire and preserve all moving images of national production . . . considered by Member States as an integral part of their 'moving image heritage'".

1.31. The Recommendation foresees a mix of governmental and non-governmental archives as achieving this objective in some countries, particularly where such non-governmental archives already exist. What practical effect this Recommendation will have remains to be seen, but its existence is an acknowledgement of a world-wide concern that moving images are a part of our global cultural heritage, in every sense of that term, which has already suffered extensive damage through neglect and acts of deliberate destruction. National archives will inevitably be called upon to play a larger role in assuring that this aspect of the historical record will be conserved but whether this will be an active role or merely one of coordination and the establishment of standards may well vary with circumstances from country to country. Certainly in countries where there is no institutionalized effort to acquire and conserve moving image documentation at present there is a strong argument for national archives to assume that responsibility, or at least to ensure that this responsibility is delegated to an organization serving the public interest.

*This section is based upon a paper prepared by the author for the Section of Professional Archives Associations, International Council on Archives, at the ICA London Congress, 1980.


Contents - Previous - Next